Mutations in the fragile X mental retardation (FMR) gene cause fragile X syndrome—the most common form of inherited mental retardation in humans. Clues into how these mutations lead to deficiencies in neuronal signaling that cause the disease have now been reported by Ed Manser of the A*STAR Institute of Medical Biology and the A*STAR–Duke-NUS Graduate Medical School Neuroscience Research Partnership in Singapore together with co-workers including Hwee-Goon Tay and Evonne Say.
Fragile X proteins ensure proper mRNA translation at neuronal synapses and contain a highly conserved region called the KH2 domain (Fig. 1) that binds RNAs and unidentified proteins. Manser and his co-workers identified the first known binding partner for fragile X proteins. They showed that a protein kinase called PAK1, which regulates the actin network and cell division, activates fragile X proteins FMR1 and FXR1 by binding to and modifying their KH2 domains.
During neural development, PAK1 is involved in the formation of synapses and dendritic spines, the tiny finger-like projections at which signaling between nerve cells takes place. Upon activation, kinases such as PAK1 assume an open configuration that exposes an amino-acid sequence that can bind to other proteins.
In a series of biochemical assays, the researchers demonstrated that the KH2 region of human FXR1 interacts with PAK1, and PAK1 modifies FXR1 upon binding by adding a phosphate group to a specific site in the KH2 domain called Ser420.
Manser’s team also developed an antibody that recognizes the ‘phosphorylated’ Ser420, and showed that active PAK1 and phosphorylated FXR1 are recruited to structures called stress granules, which form when cells are exposed to toxic chemicals.
Researchers outside of Singapore showed previously that mice and frogs lacking FXR1 have major heart and skeletal muscle defects. Manser and his co-workers found that blocking FXR1 expression with RNA inhibition produced the same defects in zebrafish embryos. They could correct the defects by injecting human FXR1 mRNA transcripts into the embryos, but not by injecting a mutant FXR1 that cannot recruit PAK1.
These results show that the FXR1/ PAK1 interaction is required for proper fragile X function in zebrafish. This is likely to be true for the proper formation of synaptic connections in the brains of zebrafish and humans.
“Many PAK1 protein targets have been identified, but few are validated,” says Manser. “In fact, we have recently discovered and purified a completely new protein.” This second binding partner for the KH2 domain of fragile X “probably works with PAK1 to modulate mRNA translation.”